Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    automatic runtime handling

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     negros
    0.35
     kabar
    0.32
     فنا
    0.32
     фено
    0.32
     discurso
    0.31
     ясно
    0.31
     христиан
    0.31
     우리나라
    0.31
     comen
    0.31
     плен
    0.30
    POSITIVE LOGITS
     dynamically
    0.59
     automatically
    0.58
     caching
    0.57
    自动
    0.55
    处理
    0.54
     runtime
    0.53
    异步
    0.53
     configurable
    0.52
     daemon
    0.52
     asynchronous
    0.50
    Act Density 2.196%

    No Known Activations